Your website form says "Thank you for your enquiry" to the customer, but that contact doesn't appear in the CRM until tomorrow morning. Nobody on sales knows why. You ask the person who set it up, and they say "it's a webhook." That tells you nothing about why the data arrived late, or how to fix it if it stops again.
So what is a webhook, and why does it matter for automation? It is an automated message one app sends to another the moment something happens. Nothing has to keep checking whether new data exists. This page explains how that differs from polling on a timer, walks through one real flow you might recognise, covers what quietly breaks when deliveries stop, and gives you the security questions to ask before trusting one.
What is a webhook in plain English?
A webhook is an automated message sent between apps, carrying detailed information about a new item. It works like a push notification rather than you going back to check: the sending app hands off the data the instant the event fires, and the receiving tool acts on it right away. You do not need to understand HTTP or JSON; what matters is that one side tells the other "this just happened" without being asked.
In any setup an owner will actually see three parts: the event (the thing that triggers the message, like a form submission or a new booking), the webhook URL (the unique address the receiving tool gives you to paste into the sending app's settings so it knows where to deliver), and the payload (the data that arrives with the message, such as the customer's name, email and phone number from the form).
You will not see most of this in a dashboard. You notice it when something goes wrong, or when you are setting up the connection for the first time and copying that URL into another app's settings page. That is why treating the webhook URL like a key matters: anyone who has it can send data to your system unless it is protected. We come back to that below.
How is a webhook different from polling?
Make describes webhooks as "instant triggers," executed immediately after the URL receives a request. Polling is the opposite. The automation tool periodically asks an app for new data, checking on a schedule whether anything has changed since last time. If your form tool or CRM does not offer webhook settings at all, polling is the fallback. Zapier calls this kind of timed check a Retrieve Poll trigger.
The practical difference shows up in speed and missed events. A webhook fires the moment the event happens; a polling check only picks up changes when its next scheduled run arrives. If you are building a daily report, that gap does not matter. But if a salesperson needs to see a new enquiry before their first coffee, it does.
You do not need to know how often your tool polls. What you should know is whether the app offers webhooks at all. If it does and you are using polling instead, you're leaving speed on the table for no reason. And if it doesn't offer them, polling works fine as long as nobody expects real-time updates from that flow.
What does a webhook look like in a real business flow?
Here's an example that will feel familiar: someone fills out your website enquiry form. The form tool has been set up with a webhook URL pointing to your automation platform. As soon as the submission lands, it fires off a message containing the name, email and any other fields they filled in. Your automation tool receives that payload, creates a contact record in your CRM, and sends a notification to the sales team's shared inbox or chat channel.
The setup is copy-and-paste. You open your automation tool, create the workflow, and it generates a unique webhook URL for you. In n8n, this shows up as two addresses: one for testing while the workflow is still being built, and one production URL that activates once you publish the flow. You copy the right one and paste it into the form tool's webhook settings. Save. Submit a test form. Check whether the contact appears in the CRM.
That's genuinely all there is to wiring up the connection on either side. The hard part isn't the setup; it's knowing that this particular URL now exists in two places. If someone moves it or deletes it, the whole flow stops silently. We cover that next. If you want to see a tool that receives webhooks and runs workflows without needing a developer, take a look at our guide to n8n.
What goes wrong when a webhook stops arriving?
This is the part most explainers skip, because it doesn't make for good documentation. The failure modes are boring and quiet:
- The Zap or scenario gets switched off. Zapier will return a 404 error to any incoming request once a Zap is turned off or deleted, but there can be a delay of up to several hours before that response takes effect. During that window, your form tool thinks the delivery succeeded. It didn't.
- Someone transfers ownership. If a Zap moves to a new owner in Zapier, the webhook URL changes. The sending app still posts to the old address. Nothing breaks loudly; it just stops working.
- The scenario goes unconnected. Make automatically deactivates webhooks that are not connected to any scenario for more than five days. You simply stop receiving data and may not notice for a while.
- Logs expire before you check. Make stores webhook logs for three days on standard plans, thirty days on Enterprise. If your problem started last Tuesday and you look at the logs next Monday, they're gone.
The fix isn't more technology; it's ownership. Someone should keep a list of every webhook URL in use, note which app sends to it and which receives from it, and do a weekly check that records are still arriving. An alert on failure helps, but the human who owns the list matters more than the automation around it.
How do you check a webhook is safe to rely on?
A webhook URL with no protection is an open door. Anyone who finds it can send data into your system. In n8n's documentation, if the IP allowlist is left blank, all IP addresses can invoke the trigger. That means a competitor, a spammer, or a misconfigured script could post fake bookings into your workflow. A small business owner on Reddit described an n8n webhook that sent details to a Telegram channel. Filtering out junk before staff see it is easier than cleaning up afterwards.
We tell clients to check who can call it (is there authentication or an IP allowlist?), whether the delivery is signed (so the receiver can verify it came from the real sender and wasn't tampered with in transit), where the secret is stored, who gets told when it fails, and who owns it. The signing token should be a random string with high entropy, kept out of code repositories and shared documents. One person per webhook URL, written down, reviewed quarterly.
If the signing token is pasted into a Slack message or a Notion page, it might as well not exist. GitHub's documentation explains that you validate a webhook's signature before processing it, to confirm both origin and integrity. Ask your supplier or developer whether this is enabled. Someone needs to know when it fails, within hours rather than weeks.
Does your business need webhooks?
You probably do if any of these sound familiar: staff copy data by hand between a form, a CRM and a spreadsheet every morning; enquiries from your booking system wait for a scheduled sync before anyone sees them; or a task should start the moment something happens (a payment clears, an order ships, a lead registers) rather than at the next half-hourly check.
One small business owner on Reddit asked for systems that "work well with automation (API, webhooks, Zapier/Make)" when evaluating new software. That phrasing tells you something: they know they want the data to move automatically and are asking whether a given tool can do it. If your current stack cannot push data out without being polled, or if the polling interval is longer than your customers' patience, webhooks solve that gap.
You don't need them when a daily report is genuinely fine, or when two apps already have a native integration that syncs in near-real-time without extra tooling. Don't add complexity to solve a problem you don't have.
Does this apply to you?
Probably not yet
With none or one of these, a clear list of where data is copied by hand is the right first step.
What should you do next?
Start with the list. Open your notes app and write down every place where data currently moves by hand or on a schedule: which form feeds which CRM, which booking tool pushes to which email address, which spreadsheet gets updated each morning. For each one, check whether the sending app has webhook settings in its integration panel or help pages. If it does and you're not using them, that's your first candidate flow.
Pick one. Build the smallest version: form to CRM contact, with a notification to whoever needs to know. Test it with a real submission from a colleague, confirm the data lands correctly, then add the monitoring: an alert if no deliveries arrive within your expected window, and a named owner for that URL. Once that's running and trusted, add the next flow.
If you want to see how this plays out in practice, read our workflow automation case study or browse our Zapier vs n8n comparison if you're choosing between the two tools for handling webhooks. And if you'd rather skip the setup and the monitoring altogether, that's what our automation service is for: we build it, test it, name the owner, and keep it running so a missed webhook is caught sooner.
In practice · Case study
Workflow Automation
Giving the editorial team their time back — permanently.
Read the case studyKey terms
- webhook
- An automated message one app sends to another when something happens, carrying details of the new item. The receiving app acts on it straight away.
- webhook URL
- The unique address the receiving tool gives you. You paste it into the sending app's settings so it knows where to deliver each message.
- polling
- Checking for new data on a schedule. The automation tool asks the app whether anything has changed, rather than being told straight away.
- trigger
- The event that starts a workflow. In an automation tool, a trigger might be a form submission, a new booking or an incoming webhook message.
- payload
- The data that arrives with a webhook message, such as a customer's name, email address and phone number.
- signature
- A check that a delivery came from the real sender and was not changed in transit. The sender adds a signature and the receiver validates it.
Frequently asked questions
What is a webhook URL?
Do I need to be a developer to use a webhook?
What happens if a webhook fails or goes missing?
Which should I choose, a webhook or polling?
How do I stop spam or fake data arriving through a webhook?
Sources
- Webhooks, Make (accessed 4 Nov 2026)
- Trigger Zap workflows from webhooks, Zapier (accessed 4 Nov 2026)
- Webhook node, n8n (accessed 4 Nov 2026)
- Validating webhook deliveries, GitHub Docs (accessed 4 Nov 2026)